Back to skill

Security audit

Pipeworx weather

Security checks for vulnerabilities and agentic risk

Overview

This is a simple weather skill that sends coordinates to a disclosed remote weather MCP endpoint and does not show hidden code, persistence, credential access, or destructive behavior.

Before installing, understand that weather lookups will send the latitude and longitude you provide to gateway.pipeworx.io, so avoid using highly sensitive exact locations unless you trust that service. Treat returned weather data as remote service output, not as instructions for the agent.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (1)

T08 · Insecure Dependencies

Warning
Location
SKILL.md:14
Finding

Unverified Third-Party Weather Gateway

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 14–16; related MCP configuration at lines 34–37
Vulnerability Type: Insecure remote service dependency
Risk Level: Medium

Vulnerable Code

bash
curl -X POST https://gateway.pipeworx.io/weather/mcp \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"get_weather","arguments":{"latitude":37.7749,"longitude":-122.4194}}}'

The same intermediary is configured as the MCP server:

json
{
  "mcpServers": {
    "weather": {
      "url": "https://gateway.pipeworx.io/weather/mcp"
    }
  }
}

Technical Analysis

The Skill instructs users and agents to submit weather requests to gateway.pipeworx.io, while separately stating that the service is powered by Open-Meteo. Requests are therefore routed through a third-party intermediary rather than directly to an official Open-Meteo endpoint.

The project contains no source code for the gateway, integrity verification mechanism, response authentication, privacy policy, ownership verification, or documented trust justification. Consequently, the effective behavior of the remote dependency can change without any modification to the reviewed Skill. Although HTTPS protects traffic in transit against ordinary network interception, it does not protect against a malicious, compromised, or incorrectly operated gateway.

This is a supply-chain and trust-boundary risk. Weather responses returned by the gateway must be treated as untrusted remote input.

Attack Path

  1. A user installs or adopts the MCP configuration from SKILL.md.
  2. The agent sends latitude, longitude, requested forecast duration, and JSON-RPC metadata to gateway.pipeworx.io.
  3. The gateway operator—or an attacker who compromises the gateway—records the submitted location queries.
  4. The gateway can return altered, fabricated, stale, or otherwise attacker-co ...[truncated 1159 chars]
Remediation
View remediation

Remediation Suggestions

  1. Replace the intermediary with the documented official Open-Meteo HTTPS API where practical.
  2. If the intermediary is required, document its ownership, operational purpose, privacy and retention practices, security controls, and relationship with Open-Meteo.
  3. Minimize submitted location precision when exact coordinates are unnecessary and clearly disclose that coordinates are sent to a third party.
  4. Validate response structure, types, ranges, and forecast-day limits before using returned data.
  5. Treat all response fields as untrusted input and prevent them from being interpreted as agent instructions, commands, or executable content.
  6. Apply request timeouts, response-size limits, retry limits, and graceful failure handling.
  7. For security-sensitive workflows, authenticate responses or cross-check consequential weather results against an independent trusted source.
  8. Pin or otherwise govern the remote service contract so unexpected changes are detected and reviewed.
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (3)

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 16)May include surrounding context.

Example: current weather in San Francisco

bash
curl -X POST https://gateway.pipeworx.io/weather/mcp \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"get_weather","arguments":{"latitude":37.7749,"longitude":-122.4194}}}'

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
91% confidence
Finding

The description states 'Temperatures in Fahrenheit, wind in mph,' which imposes a specific locale/unit convention in natural language. There is no indication that users can opt into other units or that the skill is intentionally limited to a region where imperial units are required.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
87% confidence
Finding

The skill description includes curl examples that send user-supplied location data to https://gateway.pipeworx.io/weather/mcp, but the markdown does not explicitly warn that queries are transmitted to an external service. For a user-facing skill description, a brief disclosure about external network use and shared location data would improve safety transparency.

Content

No source excerpt is available for this finding.